Willder
Shared, permissioned memory for coding agents, with an audit row per read
Emerged finds new products worth a look, daily — what’s new today
FastRecall
Hosted conversation memory that follows you from one AI model to the next
If you route between models, it keeps each conversation's context in one place and charges for what you store rather than for every recall.
FastRecall is a hosted context store for apps that switch between AI models. Before each turn you call a recall endpoint and get back a messages array to send straight to your provider; afterwards you store the reply. It hosts no models, sits alongside provider prompt caching, and can compact long contexts when full fidelity is not needed. There is a REST API with JavaScript and Python examples, a one-week sandbox with no card, and metered plans from $2, $7 and $25 a month with recalls included. The founder says he worked on memory at OpenAI. No benchmarks are published yet.
More like this
Strata
Shared memory for Claude Code and Codex sessions, with an LLM judge at the door
agent-memory
Plain-Markdown long-term memory shared by Claude Code and Codex
DaiDocs
An open plain-text file format for AI memory that any model can read
Aura
An agent appliance you host, with one memory database per person
Fraise
A single-binary memory database for AI agents with a two-verb query language